Overview / Usage

It is a scientific research using anonymous institutional data to subsidize the execution of the dissertation work in the Master Course in Computer Science of the Federal University of Alagoas (UFAL), under the supervision of Professor Dr. Thales Miranda de Almeida Vieira and coorientation of Professor Dr. Evandro de Barros Costa initially titled: Development of a model based on deep learning for the creation of predictors of absence due to illness of public security professionals in the Military Police of Alagoas.
The research aims at the development of a model based on Deep Learning (Artificial Neural Networks) for the creation of a technological tool capable of learning based on the history of departures due to illness of Public Safety professionals correlating them with the characteristics of the work activities which resulted in departures exceeding 30 (thirty) days per year.
The algorithm that will be developed will help the Military Police of Alagoas to elaborate models of groups of risks that would be automatically selected by the system to be submitted to preventive actions, subsidizing the activities of the Health professionals of the Corporation by providing an unpublished tool and endowed with the more modern implementations available for the monitoring of the health of the Military Police.
The project aims to complement the work already carried out by the Research on Health, Safety at Work and Quality of Life of Public Safety Operators of Alagoas, entitled Voices and Senses of Work of the Public Safety Operators of the State of Alagoas, executed by UFAL in the year 2015, having as contracting at the time the Secretary of State for Social Defense, currently Secretary of Public Security.
The data to be used are the unidentified records of departures from military police due to illness and information regarding the workload performed by them in subsequent annual periods, their hierarchical levels and information on the units where they are allocated. Therefore, in order to base the research, it will be necessary to carry out a statistical survey on the removals due to illness in the Corporation, reinforcing that there will be no identification of any military and the information collected will be restricted to those already published daily in the bulletins of the Corporation.
Interviews will also be conducted with the Officers of the Health Board of the Corporation to survey the needs of those professionals and to acquire knowledge with specialists in the area. I emphasize that all material collected will be submitted to the Research Ethics Committee of the Federal University of Alagoas to ensure its compliance with current legislation and ethical precepts of research widely adopted.

Technologies Used

Tensorflow, Keras, Intel Cloud, Pandas, Python, Multi Layer Perceptron